A method and system for analysis of a facility may include providing an emulation host system, generating a pristine circuit model on the emulation host system, inserting a first hardware trojan model, emulating operation of the golden circuit model, and emulating operation of the first hardware trojan model, and determine a set of machine-learning models, detecting the presence of an unknown trojan as a function of the set of machine learning models and using the same to authenticate the integrity of a GPS signal.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method of authenticating the integrity of a GPS signal, said method comprising the steps of: operating an electronic facility comprising a plurality of ML instruments, said plurality of ML instruments being adapted to store operational navigation data and GPS waypoint data; receiving said operational navigation data and said GPS waypoint data from said plurality of ML instruments; receiving a trusted waypoint; receiving a normal signature; calculating an operational waypoint as a function of said trusted waypoint and said operational navigation data; detecting anomalous behavior of said operational navigation data as a function of said normal signature and said operational navigation data; determining if the operational waypoint data and the GPS waypoint data are substantially equivalent; and authenticating said GPS waypoint data as a function of said error factor and said detection of anomalous behavior.
The invention relates to a method for verifying the integrity of GPS signals using machine learning (ML) instruments. The problem addressed is ensuring the reliability of GPS waypoint data, which can be compromised by spoofing or other errors. The method involves an electronic facility equipped with multiple ML instruments that store operational navigation data and GPS waypoint data. These instruments collect and process the data, which is then compared against a trusted waypoint and a normal signature—a baseline for expected behavior. The system calculates an operational waypoint by combining the trusted waypoint with the operational navigation data. It then detects anomalies in the operational navigation data by comparing it to the normal signature. The method checks whether the operational waypoint and the GPS waypoint data are substantially equivalent. If the data aligns and no anomalies are detected, the GPS waypoint data is authenticated. The authentication process considers both the equivalence of the waypoints and the absence of anomalous behavior, ensuring the integrity of the GPS signal. This approach enhances the reliability of GPS navigation by leveraging ML-based anomaly detection and trusted reference points.
2. The method of claim 1 wherein said substantial equivalence is further characterized as within an error bound.
A system and method for determining substantial equivalence between two or more data sets, particularly in applications requiring precise comparison such as data validation, error detection, or system synchronization. The method involves comparing the data sets to assess whether they are substantially equivalent, meaning they match within a defined tolerance or error bound. This error bound ensures that minor discrepancies, such as those caused by rounding, transmission errors, or computational noise, do not falsely indicate a lack of equivalence. The comparison process may involve statistical analysis, checksum verification, or other techniques to quantify the degree of similarity between the data sets. The error bound is dynamically adjustable based on the application requirements, allowing for flexibility in determining what constitutes an acceptable level of equivalence. This method is particularly useful in fields like financial transactions, scientific data analysis, and real-time system monitoring, where precise and reliable data comparison is critical. The system may include preprocessing steps to normalize or filter the data before comparison, further enhancing accuracy. The method ensures that only data sets meeting the specified equivalence criteria are deemed valid, reducing the risk of errors in downstream processes.
3. The method of claim 1 wherein said operational navigation data if further characterized as comprising throttle data, steering data, and compass data.
This invention relates to systems for collecting and analyzing operational navigation data from vehicles, particularly for improving navigation accuracy and performance. The technology addresses the challenge of accurately tracking and recording vehicle movements, which is critical for applications such as autonomous driving, fleet management, and navigation system calibration. The invention focuses on enhancing the granularity and reliability of navigation data by incorporating specific types of operational inputs. The method involves capturing and processing navigation data from a vehicle, with a particular emphasis on throttle data, steering data, and compass data. Throttle data provides information on the vehicle's acceleration and deceleration, while steering data tracks the direction and angle of the vehicle's movement. Compass data offers orientation information, ensuring accurate directional tracking. By integrating these data types, the system can generate a more comprehensive and precise navigation profile. This detailed data collection allows for better analysis of vehicle behavior, improved route planning, and enhanced safety features. The method may also include additional data sources, such as GPS or inertial measurement units, to further refine the navigation model. The overall goal is to provide a robust and reliable navigation solution that adapts to real-world driving conditions.
4. The method of claim 1 wherein said detecting anomalous behavior is performed inside of the ML instrument.
The invention relates to a system for detecting anomalous behavior within a machine learning (ML) instrument. The problem addressed is the need for real-time, on-device detection of irregularities in ML model performance, reducing reliance on external monitoring systems. The system includes an ML instrument that processes input data to generate outputs, where the instrument itself monitors its own behavior for deviations from expected patterns. This involves analyzing the instrument's internal operations, such as data processing steps, model predictions, or system metrics, to identify anomalies that may indicate errors, attacks, or performance degradation. The detection is performed entirely within the ML instrument, ensuring low-latency responses and enhanced security by minimizing exposure of sensitive data to external systems. The method may include comparing current behavior against predefined thresholds, statistical baselines, or learned models of normal operation. By integrating detection directly into the ML instrument, the system improves reliability and reduces the risk of undetected failures or malicious activity. The approach is particularly useful in environments where real-time monitoring is critical, such as autonomous systems, cybersecurity applications, or industrial automation.
5. An electronic facility configure to perform the method of claim 1 .
This invention relates to an electronic facility designed to process and analyze data using a specific method. The facility is configured to receive input data, which may include sensor readings, user inputs, or other forms of digital information. The system then applies a predefined algorithm to transform or analyze the input data, producing an output that may be used for decision-making, monitoring, or further processing. The algorithm may involve mathematical operations, statistical analysis, machine learning techniques, or other computational processes. The facility is also capable of storing the input data, intermediate results, and final outputs in a structured format for retrieval and review. Additionally, the system may include user interfaces or communication modules to interact with external devices or systems, allowing for real-time data exchange or remote control. The facility is designed to operate autonomously or in conjunction with other systems, ensuring reliable and efficient data processing. The invention addresses the need for automated data analysis in various applications, such as industrial monitoring, healthcare diagnostics, or financial forecasting, by providing a scalable and adaptable electronic solution.
6. A non-transitory computer readable medium storing computer readable instructions which, when executed in a processing system, causes the processing system to perform the steps of a method according to claim 1 .
A system and method for processing data in a computing environment involves storing executable instructions on a non-transitory computer-readable medium. When executed by a processing system, these instructions cause the system to perform a series of operations. The method includes receiving input data, analyzing the data to identify relevant patterns or features, and generating an output based on the analysis. The output may include processed data, recommendations, or actions derived from the input. The system may also include additional steps such as validating the input data, applying machine learning models to the data, or optimizing the processing steps for efficiency. The method ensures accurate and efficient data handling, addressing challenges related to data processing speed, accuracy, and resource utilization in computing systems. The solution is applicable in various domains, including data analytics, automation, and decision-making systems, where reliable and efficient data processing is critical. The system improves upon existing methods by integrating advanced techniques to enhance performance and reduce computational overhead.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
December 27, 2020
March 15, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.